In October 2020, KDIGO (Kidney Disease: Improving Global Outcomes) published its first clinical practice guideline directed specifically to the care of patients with diabetes and chronic kidney disease (CKD). This commentary presents the views of the KDOQI (Kidney Disease Outcomes Quality Initiative) work group for diabetes in CKD, convened by the National Kidney Foundation to provide an independent expert perspective on the new guideline. The KDOQI work group believes that the KDIGO guideline takes a major step forward in clarifying glycemic targets and use of specific antihyperglycemic agents in diabetes and CKD. The purpose of this commentary is to carry forward the conversation regarding optimization of care for patients with diabetes and CKD. Recent developments for prevention of CKD progression and cardiovascular events in people with diabetes and CKD, particularly related to sodium/glucose cotransporter 2 (SGLT2) inhibitors, have filled a longstanding gap in nephrology's approach to the care of persons with diabetes and CKD. The multifaceted benefits of SGLT2 inhibitors have facilitated interactions between nephrology, cardiology, endocrinology, and primary care, underscoring the need for innovative approaches to multidisciplinary care in these patients. We now have more interventions to slow kidney disease progression and prevent or delay kidney failure in patients with diabetes and kidney disease, but methods to streamline their implementation and overcome barriers in access to care, particularly cost, are essential to ensuring all patients may benefit.

Background

The growing amount of high dimensional biomolecular data has spawned new statistical and computational models for risk prediction and disease classification. Yet, many of these methods do not yield biologically interpretable models, despite offering high classification accuracy. An exception, the top-scoring pair (TSP) algorithm derives parameter-free, biologically interpretable single pair decision rules that are accurate and robust in disease classification. However, standard TSP methods do not accommodate covariates that could heavily influence feature selection for the top-scoring pair. Herein, we propose a covariate-adjusted TSP method, which uses residuals from a regression of features on the covariates for identifying top scoring pairs. We conduct simulations and a data application to investigate our method, and compare it to existing classifiers, LASSO and random forests.

Results

Our simulations found that features that were highly correlated with clinical variables had high likelihood of being selected as top scoring pairs in the standard TSP setting. However, through residualization, our covariate-adjusted TSP was able to identify new top scoring pairs, that were largely uncorrelated with clinical variables. In the data application, using patients with diabetes (n = 977) selected for metabolomic profiling in the Chronic Renal Insufficiency Cohort (CRIC) study, the standard TSP algorithm identified (valine-betaine, dimethyl-arg) as the top-scoring metabolite pair for classifying diabetic kidney disease (DKD) severity, whereas the covariate-adjusted TSP method identified the pair (pipazethate, octaethylene glycol) as top-scoring. Valine-betaine and dimethyl-arg had, respectively, ≥ 0.4 absolute correlation with urine albumin and serum creatinine, known prognosticators of DKD. Thus without covariate-adjustment the top-scoring pair largely reflected known markers of disease severity, whereas covariate-adjusted TSP uncovered features liberated from confounding, and identified independent prognostic markers of DKD severity. Furthermore, TSP-based methods achieved competitive classification accuracy in DKD to LASSO and random forests, while providing more parsimonious models.

Conclusions

We extended TSP-based methods to account for covariates, via a simple, easy to implement residualizing process. Our covariate-adjusted TSP method identified metabolite features, uncorrelated from clinical covariates, that discriminate DKD severity stage based on the relative ordering between two features, and thus provide insights into future studies on the order reversals in early vs advanced disease states.

INTRODUCTION:Little is known about the association of urine metabolites with structural lesions in persons with diabetes. OBJECTIVES:We examined the relationship between 12 urine metabolites and kidney structure in American Indians with type 2 diabetes. METHODS:Data were from a 6-year clinical trial that assessed renoprotective efficacy of losartan, and included a kidney biopsy at the end of the treatment period. Metabolites were measured in urine samples collected within a median of 6.5 months before the research biopsy. Associations of the creatinine-adjusted urine metabolites with kidney structural variables were examined by Pearson's correlations and multivariable linear regression after adjustment for age, sex, diabetes duration, hemoglobin A1c, mean arterial pressure, glomerular filtration rate (iothalamate), and losartan treatment. RESULTS:Participants (n = 62, mean age 45 ± 10 years) had mean ± standard deviation glomerular filtration rate of 137 ± 50 ml/min and median (interquartile range) urine albumin:creatinine ratio of 34 (14-85) mg/g near the time of the biopsy. Urine aconitic and glycolic acids correlated positively with glomerular filtration surface density (partial r = 0.29, P = 0.030 and r = 0.50, P < 0.001) and total filtration surface per glomerulus (partial r = 0.32, P = 0.019 and r = 0.43, P = 0.001). 2-ethyl 3-OH propionate correlated positively with the percentage of fenestrated endothelium (partial r = 0.32, P = 0.019). Citric acid correlated negatively with mesangial fractional volume (partial r=-0.36, P = 0.007), and homovanillic acid correlated negatively with podocyte foot process width (partial r=-0.31, P = 0.022). CONCLUSIONS:Alterations of urine metabolites may associate with early glomerular lesions in diabetic kidney disease.

© 2015. Background: A growing number of serum filtration markers are associated with mortality and end-stage renal disease (ESRD) in adults. Whether β-trace protein (BTP) and β2-microglobulin (B2M) are associated with these outcomes in adults with type 2 diabetes is not known. Study Design: Longitudinal cohort study. Setting & Participants: 250 Pima Indians with type 2 diabetes (69% women; mean age, 42 years; mean diabetes duration, 11 years). Predictors: Serum BTP, B2M, and glomerular filtration rate measured by iothalamate clearance (mGFR) or estimated using creatinine (eGFRcr) or cystatin C level (eGFRcys). Outcomes & Measurements: Incident ESRD and all-cause mortality through December 2013. HRs were reported per interquartile range decrease of the inverse of BTP and B2M (1/BTP and 1/B2M) using Cox regression. Improvement in risk prediction with the addition of BTP or B2M level to established markers (eGFRcys with mGFR or eGFRcr) was evaluated using C statistics, continuous net reclassification improvement, and relative integrated discrimination improvement (RIDI). Results: During a median follow-up of 14 years, 69 participants developed ESRD and 95 died. Both novel markers were associated with ESRD in multivariable models. BTP level remained statistically significant after further adjustment for mGFR (1/BTP, 1.53 [95% CI, 1.01-2.30]; 1/B2M, 1.54 [95% CI, 0.98-2.42]). B2M level was associated with mortality in multivariable models and after further adjustment for mGFR (HR, 2.12; 95% CI, 1.38-3.26). The addition of B2M level to established markers increased the C statistic for mortality but only weakly when assessed by either continuous net reclassification improvement or RIDI; none was improved for ESRD by the addition of these markers. Limitations: Small sample size, single measurements of markers. Conclusions: In Pima Indians with type 2 diabetes, BTP and, to a lesser extent, B2M levels were associated with ESRD. B2M level was associated with mortality after adjustment for traditional risk factors and established filtration markers. Further studies are warranted to confirm whether inclusion of B2M level in a multimarker approach leads to improved risk prediction for mortality in this population.

Background

The primary biomarkers used to define CKD are serum creatinine and albuminuria. These biomarkers have directed focus on the filtration and barrier functions of the kidney glomerulus even though albuminuria results from tubule dysfunction as well. Given that proximal tubules make up ∼90% of kidney cortical mass, we evaluated whether a sensitive and specific marker of proximal tubule injury, urinary kidney injury molecule-1 (KIM-1), is elevated in individuals with CKD or with risk factors for CKD.

Methods

We measured urinary KIM-1 in participants of five cohort studies from the USA and Sweden. Participants had a wide range of kidney function and were racially and ethnically diverse. Multivariable linear regression models were used to test the association of urinary KIM-1 with demographic, clinical and laboratory values.

Results

In pooled, multivariable-adjusted analyses, log-transformed, creatinine-normalized urinary KIM-1 levels were higher in those with lower eGFR {β = -0.03 per 10 mL/min/1.73 m(2) [95% confidence interval (CI) -0.05 to -0.02]} and greater albuminuria [β = 0.16 per unit of log albumin:creatinine ratio (95% CI 0.15-0.17)]. Urinary KIM-1 levels were higher in current smokers, lower in blacks than nonblacks and lower in users versus nonusers of angiotensin-converting enzyme inhibitors and angiotensin receptor blockers.

Conclusion

Proximal tubule injury appears to be an integral and measurable element of multiple stages of CKD.

Few investigations have evaluated the incremental usefulness of tubular injury biomarkers for improved prediction of chronic kidney disease (CKD) progression. As such, we measured urinary kidney injury molecule-1, neutrophil gelatinase-associated lipocalin, N-acetyl-ß-D-glucosaminidase and liver fatty acid binding protein under highly standardized conditions among 2466 enrollees of the prospective Chronic Renal Insufficiency Cohort Study. During 9433 person-years of follow-up, there were 581 cases of CKD progression defined as incident end-stage renal disease or halving of the estimated glomerular filtration rate. Levels of the urine injury biomarkers, normalized for urine creatinine, were strongly associated with CKD progression in unadjusted Cox proportional hazard models with hazard ratios in the range of 7 to 15 comparing the highest with the lowest quintiles. However, after controlling for the serum creatinine-based estimated glomerular filtration rate and urinary albumin/creatinine ratio, none of the normalized biomarkers was independently associated with CKD progression. None of the biomarkers improved on the high (0.89) C-statistic for the base clinical model. Thus, among patients with CKD, risk prediction with a clinical model that includes the serum creatinine-based estimated glomerular filtration rate and the urinary albumin/creatinine ratio is not improved on with the addition of renal tubular injury biomarkers.

Cardiovascular disease risk factor control as primary prevention in patients with type 2 diabetes mellitus has changed substantially in the past few years. The purpose of this scientific statement is to review the current literature and key clinical trials pertaining to blood pressure and blood glucose control, cholesterol management, aspirin therapy, and lifestyle modification. We present a synthesis of the recent literature, new guidelines, and clinical targets, including screening for kidney and subclinical cardiovascular disease for the contemporary management of patients with type 2 diabetes mellitus.

Aims/hypothesis

Kidney injury molecule 1 (KIM-1), liver fatty acid-binding protein (L-FABP), N-acetyl-β-D-glucosaminidase (NAG) and neutrophil gelatinase-associated lipocalin (NGAL) are urinary biomarkers of renal tubular injury. We examined their association with incident end-stage renal disease (ESRD) and all-cause mortality in American Indians with type 2 diabetes.

Methods

Biomarker concentrations were measured in baseline urine samples in 260 Pima Indians who were followed for a median of 14 years. HRs were reported per SD of creatinine (Cr)-normalised log-transformed KIM-1, NAG and NGAL, and for three categories of L-FABP.

Results

During follow-up, 74 participants developed ESRD and 101 died. Median concentrations of KIM-1/Cr, NAG/Cr and NGAL/Cr and the proportion of detectable L-FABP were highest in those with macroalbuminuria (p < 0.001 for KIM-1/Cr, NAG/Cr and L-FABP; p = 0.006 for NGAL/Cr). After multivariable adjustment, NGAL/Cr was positively associated with ESRD (HR 1.59, 95% CI 1.20, 2.11) and mortality (HR 1.39, 95% CI 1.06, 1.82); L-FABP/Cr was inversely associated with ESRD (HR [for highest vs lowest tertile] 0.40, 95% CI 0.19, 0.83). Addition of NGAL/Cr to models that included albuminuria and glomerular filtration rate increased the c-statistic for predicting ESRD from 0.828 to 0.833 (p = 0.001) and for death from 0.710 to 0.722 (p = 0.018). Addition of L-FABP/Cr increased the c-statistic for ESRD from 0.828 to 0.832 (p = 0.042).

Conclusions/interpretation

In Pima Indians with type 2 diabetes, urinary concentrations of NGAL and L-FABP are associated with important health outcomes, but they are unlikely to add to risk prediction with standard markers in a clinically meaningful way given the small increase in the c-statistic.

Rationale & objective

Plasma kidney injury molecule 1 (KIM-1) is a sensitive marker of proximal tubule injury, but its association with risks of adverse clinical outcomes across a spectrum of kidney diseases is unknown.

Study design

Prospective, observational cohort study.

Setting & participants

524 individuals enrolled into the Boston Kidney Biopsy Cohort (BKBC) Study undergoing clinically indicated native kidney biopsy with biopsy specimens adjudicated for semiquantitative scores of histopathology by 2 kidney pathologists and 3,800 individuals with common forms of chronic kidney disease (CKD) enrolled into the Chronic Renal Insufficiency Cohort (CRIC) Study.

Exposure

Histopathologic lesions and clinicopathologic diagnosis in cross-sectional analyses, baseline plasma KIM-1 levels in prospective analyses.

Outcomes

Baseline plasma KIM-1 levels in cross-sectional analyses, kidney failure (defined as initiation of kidney replacement therapy) and death in prospective analyses.

Analytical approach

Multivariable-adjusted linear regression models tested associations of plasma KIM-1 levels with histopathologic lesions and clinicopathologic diagnoses. Cox proportional hazards models tested associations of plasma KIM-1 levels with future kidney failure and death.

Results

In the BKBC Study, higher plasma KIM-1 levels were associated with more severe acute tubular injury, tubulointerstitial inflammation, and more severe mesangial expansion after multivariable adjustment. Participants with diabetic nephropathy, glomerulopathies, and tubulointerstitial disease had significantly higher plasma KIM-1 levels after multivariable adjustment. In the BKBC Study, CKD in 124 participants progressed to kidney failure and 85 participants died during a median follow-up time of 5 years. In the CRIC Study, CKD in 1,153 participants progressed to kidney failure and 1,356 participants died during a median follow-up time of 11.5 years. In both cohorts, each doubling of plasma KIM-1 level was associated with an increased risk of kidney failure after multivariable adjustment (hazard ratios of 1.19 [95% CI, 1.03-1.38] and 1.10 [95% CI, 1.06-1.15] for BKBC and CRIC, respectively). There was no statistically significant association of plasma KIM-1 levels with death in either cohort.

Limitations

Generalizability and unmeasured confounding.

Conclusions

Plasma KIM-1 is associated with underlying tubulointerstitial and mesangial lesions and progression to kidney failure in 2 cohort studies of individuals with kidney diseases.